Combination of chemical genetics and phosphoproteomics for kinase signaling analysis enables confident identification of cellular downstream targets

Abstract

Delineation of phosphorylation-based signaling networks requires reliable data about the underlying cellular kinase-substrate interactions. We report a chemical genetics and quantitative phosphoproteomics approach that encompasses cellular kinase activation in combination with comparative replicate mass spectrometry analyses of cells expressing either inhibitor-sensitive or resistant kinase variant. We applied this workflow to Plk1 (Polo-like kinase 1) in mitotic cells and induced cellular Plk1 activity by wash-out of the bulky kinase inhibitor 3-MB-PP1, which targets a mutant kinase version with an enlarged catalytic pocket while not interfering with wild-type Plk1. We quantified more than 20,000 distinct phosphorylation sites by SILAC, approximately half of which were measured in at least two independent experiments in cells expressing mutant and wild-type Plk1. Based on replicate phosphorylation site quantifications in both mutant and wild-type Plk1 cells, our chemical genetic proteomics concept enabled stringent comparative statistics by significance analysis of microarrays, which unveiled more than 350 cellular downstream targets of Plk1 validated by full concordance of both statistical and experimental data. Our data point to hitherto poorly characterized aspects in Plk1-controlled mitotic progression and provide a largely extended resource for functional studies. We anticipate the described strategies to be of general utility for systematic and confident identification of cellular protein kinase substrates.

Fig. 1.

Combined chemical genetics and phosphoproteomics approach to identify cellular downstream targets of protein kinases. A, cellular induction of Plk1 activity. Plk1^as cells were synchronized and then arrested in mitosis by either nocodazole treatment (N) or nocodazole plus 5 μm 3-MB-PP1 inhibitor treatment (N+I) for 13 h. The cells were then lysed (—) or washed and incubated for a further 30 min in the presence of either nocodazole (N) or nocodazole and 3-MB-PP1 (N+I) before lysis. In addition, lysate from nonsynchronized cells was prepared (Ø). Total extracts were then analyzed by immunoblotting with BubR1-specific antibody. The slower migrating band in the second lane from the right indicates the appearance of Plk1-phosphorylated BubR1 upon 30 min of 3-MB-PP1 wash-out. This treatment condition was compared with continued nocodazole and 3-MB-PP1 incubation (third lane from the right) in all subsequent SILAC MS experiments. B, distributions of phosphorylation site ratios measured upon 3-MB-PP1 wash-out in Plk1^as and Plk1^wt cell phosphoproteomics experiments. C, scheme illustrating the overall experimental design and workflow. Four biological replicate experiments were performed in both Plk1^as and Plk1^wt cells upon SILAC encoding as indicated. In each experiment, 3-MB-PP1 was washed out from one mitotically arrested cell population before lysis and protein digestion. Phosphopeptides were enriched by SCX chromatography combined with IMAC prior to LC-MS analysis and data processing. D, comparison of Plk1^as and Plk1^wt cell experiments with respect to the overall numbers of quantified phosphorylation sites and phosphoproteins.

Fig. 2.

Evaluation of biological replicate approach for Plk1 downstream target identification. A, ratio distribution and reproducibility across replicate experiments. The distribution of phosphosite ratios for 3-MB-PP1 versus control treatment is shown as a box plot for all individual Plk1^as and Plk1^wt biological replicate experiments. For each SILAC analysis, the numbers of all quantified phosphosites and of those measured with more than 2-fold changes upon 3-MB-PP1 wash-out are shown. In the lower part, numbers of phosphosites are shown that are more than 2-fold regulated in the same direction (up- or down-regulation) for all pair-wise comparisons of either Plk1^as or Plk1^wt experiments. In addition, the corresponding numbers of all phosphosites detected in the compared experiments are indicated. B, numbers of phosphosites quantified in at least two independent experiments for two, three, and four replicate analyses in either Plk1^as or Plk1^wt cells. The average numbers determined from all possible combinations of two or three SILAC experiments are shown. C, matrix showing the numbers of phosphosites quantified in x biological replicate analyses of Plk^as and y biological replicate analyses Plk^wt cells (x, y = 0, 1, 2, 3, and 4). Phosphosite numbers for all possible combinations are shown. Phosphosites quantified in at least two Plk^as and at least two Plk^wt cell experiments are boxed.

Fig. 3.

Plk1-specific phosphoregulation. A, identification of Plk1-dependent phosphorylation changes by SAM. Shown is a scatter plot of the observed score versus the expected score showing the results of differential SAM analysis of Plk1^as or Plk1^wt ratios. The solid line indicates identity of observed and expected scores, whereas the dashed lines depict thresholds of Δ = 2.136 beyond which phosphosites were identified as significantly up- and down-regulated (red and blue circles) according to an FDR of 0%. B, regulated phosphosites with significantly different ±3-MB-PP1 ratios in Plk1^as compared with Plk1^wt cells. Average ratios from Plk1^as cell SILAC analyses were plotted against the respective ratios from Plk1^wt experiments. Selectively up- and down-regulated sites in Plk1^as versus Plk1^wt cells are shown as red and blue circles, respectively.

Fig. 4.

Enriched sequence motifs in Plk1-induced phosphorylation sites. A, phosphorylation motifs were extracted with the Motif-X algorithm (24) using all class I serine phosphorylation sites significantly up-regulated upon cellular Plk1 activation. Additional information is provided in supplemental Fig. 2. B, frequencies of the Plk1 consensus motif (D/E/N)X(pS/pT)-ϕ, where X represents any amino acid, and ϕ denotes a hydrophobic residue (25), its reduced version (D/E/N)X(pS/pT) without a hydrophobic residue in +1 position, and the Polo box domain recognition motif S(pS/pT)(P/X) motif in all proteins with induced phosphosites upon Plk1 activation compared with all phosphoproteins in our SAM analysis.

Fig. 5.

Plk1 regulation in cellular compartments and mTOR signaling. A, cellular localization of Plk1-regulated phosphoproteins. Selected GO cellular component terms are shown that were significantly over-represented for Plk1-regulated phosphoproteins compared with all proteins quantified in Plk1^as cells (p < 0.05). A full list of all GO terms, as well as their enrichment compared with all entries in the human IPI database, is provided in supplemental Table 4. The ratios represent the numbers of proteins with the indicated GO annotation divided by the number of all GO annotated proteins in the respective categories. B, Plk1 regulation of the mTOR pathway. Components of the mTOR reference pathway (from the Kyoto Encyclopedia of Genes and Genomes) are shown with Plk1-regulated proteins highlighted in red, all other identified phosphoproteins depicted in orange, and additional pathway members in blue. PRAS40 has been further added based on literature evidence (40).